Single-Crystal X-Ray Diffraction of Grunerite up to 25.6 Gpa: a New High-Pressure Clinoamphibole Polymorph

Physics and Chemistry of Minerals (2019) 46:215–227 https://doi.org/10.1007/s00269-018-0999-1

ORIGINAL PAPER

Single-crystal X-ray diffraction of grunerite up to 25.6 GPa: a new high-pressure clinoamphibole polymorph

Tommy Yong1,2 · Przemyslaw Dera1,2 · Dongzhou Zhang2

Received: 27 May 2018 / Accepted: 23 August 2018 / Published online: 5 September 2018 © Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract High-pressure single-crystal X-ray diffraction experiments were conducted on natural grunerite crystals with composition (Fe 5.237Mg1.646Ca0.061Mn0.051Na0.015Ti0.002Cr0.001K0.001)(Si7.932Al0.083)O22(OH)2, using a synchrotron X-ray source. Grunerite has C2/m symmetry at ambient conditions. The samples were compressed at 298 K in a diamond-anvil cell to a maximum pressure of 25.6(5) GPa. We observe a previously described phase transition from C2/m (α) to P21/m (β) to take place at 7.4(1) GPa, as well as a further transition from P21/m (β) to C2/m (γ) at 19.2(3) GPa. The second-order Birch–Murnaghan 3 equation of state fit to our compressional data, yielded the values V0 = 914.7(7) Å and K0 = 78(1) GPa for α-grunerite, 3 3 V0 = 926(5) Å and K0 = 66(4) GPa for β-grunerite and V0 = 925(27) Å and K0 = 66(13) GPa for γ-grunerite. The β–γ phase transition produces a greater degree of kinking in the double silicate chains of tetrahedra accompanied by a discontinuous change in the a and c unit cell parameters and the monoclinic β angle. At 22.8(4) GPa the O5–O6–O5 kinking angle of the new high-pressure C2/m phase is 137.5(4)°, which is the lowest reported for any monoclinic amphibole. This study is the first structural report to show the existence of three polymorphs within an amphibole group mineral. The high-pressure γ-phase illustrates the parallel structural relations and phase transformation behavior of both monoclinic single and double chain silicates.

Keywords Amphibole · Phase transition · High pressure · Single-crystal X-ray diffraction · Diamond anvil cell · Synchrotron source

Introduction metamorphic rocks. Amphiboles have such unique chemistry that they are useful tools as petrogenetic indicators Amphiboles are one of the most important rock-forming (Hirschmann et al. 1994). Structural studies of the amphi- mineral groups found in the Earth’s crust and upper mantle. bole group minerals have been well documented at ambient They are widespread in altered oceanic crust and subduct- conditions, however, despite their ubiquity, relatively few ing slabs in amphibolite, greenschist, and blueschist facies structural studies have been conducted under high-pressure conditions. Previous high-pressure structural studies have been limited to pressures below 10 GPa (Comodi et al. 1991, Electronic supplementary material The online version of this 2010; Zhang et al. 1992; Yang et al. 1998; Welch et al. 2007, article (https://doi.org/10.1007/s00269-018-0999-1) contains 2011; Zanazzi et al. 2010; Nestola et al. 2012), whereas supplementary material, which is available to authorized users. higher pressure studies have only involved spectroscopic * Tommy Yong observations (Iezzi et al. 2006, 2009, 2011; Thompson et al. [email protected] 2016). The cummingtonite-grunerite solid solution is of struc- 1 Department of Geology and Geophysics, School of Ocean and Earth Science and Technology, University of Hawaii tural significance and interest as this binary join has three at Mānoa, 1680 East West Road, POST Bldg, Honolulu, different ambient structural phases, orthorhombic Pnma HI 96822, USA anthophyllite (Mg end-member), monoclinic P21/m Mg- 2 Hawaii Institute of Geophysics and Planetology, School rich cummingtonites and monoclinic C2/m grunerites. of Ocean and Earth Science and Technology, University Earlier experiments on the cummingtonite-grunerite solid of Hawaii at Mānoa, 1680 East West Road, POST Bldg, solution series have shown that grunerite [Fe-end member, Honolulu, HI 96822, USA

Vol.:(0123456789)1 3 216 Physics and Chemistry of Minerals (2019) 46:215–227

2+ Fe7Si8O22(OH)2] undergoes a phase transition from C2/m as Fe to maintain charge balance, however, the presence 3+ (α) to P21/m (β) with increasing pressure (Yang et al. 1998; of trace amounts of Fe is a possibility. Garnet, chromite, Boffa Ballaran et al. 2000), while P21/m Mg-rich cumming- albite, diopside, scapolite, sphene glass and orthoclase tonites transforms to C2/m at high temperature (Prewitt et al. standards were used. 1970). The chemical formula was calculated based on 23 O The amphibole crystal structure is characterized by dou- atoms as described by Hawthorne and Oberti (2007). This ble chains of silicate tetrahedra which extend along the [001] calculation assumes that (O, OH, F, Cl) = 2 apfu and as no crystallographic direction. A band of octahedrally coordi- F and Cl were detected from the microprobe analysis, we nated cations, designated by sites M1, M2, M3 and M4, link assumed that there were 2 apfu of OH. The results from the adjacent chains of silicate tetrahedra, T1 and T2, along the microprobe analysis are shown in Table 1. a axis. The main difference between the C2/m and P21/m structure, is that the C2/m phase contains one crystallo- Ambient‑pressure X‑ray diffraction graphically distinct O-rotated silicate chain, while the P21/m structure contains two double silicate chains, the S-rotated To characterize the ambient pressure crystal structure of A chain and O-rotated B chain [see Papike and Ross (1970) the sample used in the high-pressure experiments, a euhe- for a description of S- and O-rotated chains]. Furthermore, dral, platelet crystal, approximately 0.15 × 0.09 × 0.02 mm the M4 site cation increases its coordination from 6 to 7 after in size was selected. The crystal was mounted on a Bruker the C2/m–P21/m phase transition. The structure and crystal D8 Venture single crystal diffractometer with a Ag IµS chemical relations of amphibole group minerals have been microfocus source (0.56089 Å) and PHOTON-II CPAD well documented by Papike and Ross (1970), Papike and detector at the University of Hawaii at Manoa’s X-ray Atlas Cameron (1976), Law and Whittaker (1980) and Hawthorne Diffraction Laboratory. The X-ray diffraction data were and Oberti (2007). collected from a θ range of 3.077° to 25.547° with com- Spectroscopic experiments on a synthetic amphibole with pleteness to θ = 19.665. Least-squares structure refinement composition, Na(NaMg)Mg5Si8O22(OH)2 and P21/m ambi- was done with the program SHELXL (Sheldrick 2008). ent symmetry, have indicated a possible existence of a new The initial structure model of grunerite from Finger (1969) high-pressure phase characterized by a C-centered lattice, was used. All atoms were refined using anisotropic atomic with the phase transition likely to occur between 20 and displacement parameters. Full occupancy was assumed for 22 GPa (Iezzi et al. 2006). This new high-pressure phase the M1, M2, M3, T1 and T2 sites, based on the calculated change may be analogous to the structural changes seen chemical formula for our sample. Partial occupancy was in clinopyroxenes with increasing pressure, however, the refined for the M4 site, which is nominally occupied by previous study did not include structure determination. To Fe 2+ (Hirschmann et al. 1994). The A site was assumed to shed new light on the nature of the higher-pressure amphi- be unoccupied as stoichiometric grunerite has a vacant A bole phases we reinvestigated the compressional behavior site, despite our crystal having a small amount of Na and of natural grunerite using synchrotron-based single crystal K, which occupies the A site. The small amount of Na and X-ray diffraction experiments. In this study, we report the K in our crystal is equivalent to < 0.19 of an electron per results of a previously unknown phase transition in natural unit cell, which is too small of a charge to be detected by grunerite between 16.3(3) and 19.2(3) GPa from P21/m (β) the difference-Fourier maps in the structure refinement. The to C2/m (γ). structure was refined using the determined chemical formula from the microprobe analysis as a restraint, however, as diffraction experiments cannot resolve small compositional Experimental procedures differences, the small amounts of Ti (0.002 apfu) and Cr (0.001 apfu) were ignored. Fe and Mn have similar X-ray Chemical analysis atomic scattering factors, and as such, these atoms were grouped together in the refinement. The M1, M2 and M3 In this study, we used a natural grunerite sample from sites were only occupied by Mg2+ and Fe2+, the M4 site was Moose Mountain Mine, Ontario, Canada with composition occupied by Fe2+ and Ca2+ and the T1 and T2 sites only (Fe 5.237Mg1.646Ca0.061Mn0.051Na0.015Ti0.002Cr0.001K0.001) contained Si. The refinement assumed no substitutional dis- 2+ (Si7.932Al0.083)O22(OH)2 determined by wavelength-disper- order of Mg in the M4 site as there is a strong preference sive spectrometry (WDS) using a JEOL Hyperprobe JXA- for Fe2+ and Ca2+ in the M4 site (Hirschmann et al. 1994). 8500F at 15 keV. Composition was determined through the The determined site occupancies for the four M sites are: average of 16 spot analyses on three different single crystals. M1: Fe2+ = 0.731(4), Mg2+ = 0.269; M2: Fe2+ = 0.560(4), The samples were homogenous with no zoning as evidenced Mg 2+ = 0.440; M3: Fe2+ = 0.749(6), Mg 2+ = 0.251; M4: from the electron backscattered image. All iron was assigned Fe2+ = 0.966(4), Ca2+ = 0.022(4). Based on these values

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 217

Table 1 Results from Constituent Wt.% Range Stand. dev. Probe standarda Crystal Line Ions/formula microprobe analyses FeO 39.14 38.06–39.9 0.52 Garnet, Verma (Mn) LiF Kα 5.237 MgO 6.9 6.47–7.21 0.28 Chromite USNM 117,075 TAP Kα 1.646

Na2O 0.05 0.02–0.1 0.02 Albite, Amelia TAP Kα 0.015

Al2O3 0.44 0.34–0.6 0.07 Chromite USNM 117,075 TAP Kα 0.083

SiO2 49.58 49.15–49.93 0.22 Albite, Amelia TAP Kα 7.932 CaO 0.36 0.28–0.47 0.05 Diopside-2 (UCLA) PETH Kα 0.061 MnO 0.38 0.31–0.49 0.05 Garnet, Verma (Mn) LiF Kα 0.051 Cl 0 0–0.02 0.01 Scapolite PETH Kα 0

TiO2 0.02 0–0.05 0.02 Sphene glass LiFH Kα 0.002

K2O 0.01 0–0.02 0.01 Orthoclase (OR-1) PETH Kα 0.001

Cr2O3 0.01 0–0.02 0.01 Chromite USNM 117,075 LiFH Kα 0.001 Total 96.8

a Probe standard compositions (Wt.%)

Garnet, Verma (Mn) = SiO2: 36.88, Al2O3: 20.82, FeO: 18.04, CaO: 0.24, MnO: 24.6

Chromite USNM 117075 = Al2O3: 9.92, FeO: 13.04, MgO:15.2, MnO: 0.1, TiO2: 0.12, Cr 2O3: 60.5, NiO: 0.16, U2O3: 0.09

Albite, Amelia = SiO2: 68.75, Al2O3: 19.43, Fe2O3: 0.02, Na2O: 11.7, K2O: 0.1

Diopside-2 (UCLA) = SiO2: 55.27, Al 2O3: 0.05, FeO: 0.94, MgO: 18.29, CaO: 25.47, MnO: 0.1, Na 2O: 0.05, TiO2: 0.06

Scapolite = SiO2: 49.78, Al2O3: 25.05, FeO: 0.17, CaO: 13.58, Na2O: 5.2, K2O: 0.94, Cl: 1.43, CO 2: 2.5, + SO3: 1.32, H2O : 0.21

Sphene glass: SiO 2: 30.65, CaO: 28.6, TiO 2: 40.75

Orthoclase (OR-1): SiO2: 64.39, Al2O3: 18.58, FeO: 0.03, Na2O: 1.14, K2O: 14.92, BaO: 0.82, SrO: 0.035, + NiO: 0.03, U2O3: 0.08, SO3: 0.03, H2O : 0.08

our refined chemical formula is(Fe 5.26Mg1.669Ca0.044)(Si8) anvils and backing plates (Boehler and De Hantsetters 2004) O22(OH)2, which is in good agreement with our calculated were used in runs 1 and 3 to increase coverage of recipro- chemical formula. cal space. For run 2, standard brilliant cut diamonds anvils with 0.300 mm culets were used on asymmetric backing High‑pressure X‑ray diffraction plates (cubic boron nitride seat towards the X-ray source and tungsten carbide toward the detector). A hole, 0.210 mm High-pressure single-crystal X-ray diffraction experiments in diameter, was drilled through a 0.250 mm thick rhenium were performed at beamlines 13BM-C and 13ID-D (GSE- gasket that was preindented to 0.040 mm to act as the sample CARS) of the Advanced Photon Source (APS), Argonne chamber. Two small ruby spheres were placed in the sample National Laboratory. Three separate experiments were chamber together with the sample crystals as a pressure cali- conducted on the natural grunerite sample. Run 1 (13BM- brant. Pressure was calculated from the shift of the R1 ruby C) consisted of 5 pressure steps ranging from 1.13(2) to fluorescence line (Dewaele et al. 2008). The DAC was gas 7.4(1) GPa, Run 2 (13ID-D) consisted of 6 pressure steps loaded at the GSECARS-COMPRES facility (Rivers et al. ranging from 9.0(1) to 25.6(5) GPa and Run 3 (13BM-C) 2008) with neon as the pressure medium to ~ 1.37 GPa. After consisted of 3 pressure steps at 10.6(2), 19.2(3) and 22.8(4) gas loading the sample chamber had shrunk to ~ 0.115 mm GPa. In run 1 we observed the known phase transition from in diameter. Ruby fluorescence spectra were measured at α to β, while in run 2 we observed for the first time the novel each pressure point both before and after the X-ray data col- phase transition from β to γ. Run 3 was used to solve the new lection. Uncertainties in pressures were taken as 2% of the γ-grunerite phase. We utilized data from three different runs pressure measurement. as there was not enough coverage of reciprocal space in run High-pressure diffraction experiments conducted at 2 to solve the new structure. experimental station 13BM-C were performed using a mon- Two crystals of grunerite with approximate size of ochromatic X-ray beam with energy of 28.6 keV (0.434 Å), 0.065 × 0.030 × 0.005 mm were loaded into a 4-pin diamond- and 1 eV bandwidth, focused with a Kirkpatrick-Baez mir- anvil cell (DAC) with 400 µm culet diamonds. Each run ror system to a spot of 0.015 mm x 0.015 mm, measured utilized a different pair of crystals, however, all crystals as full width at half maximum (FWHM). The MAR165 used in this study came from the same bulk sample. Conical charge-coupled device (CCD) detector was placed roughly

1 3 218 Physics and Chemistry of Minerals (2019) 46:215–227

180 mm away from the sample, and ambient LaB 6 powder were performed following the procedure described by was used to calibrate the distance and tilting of the detector. Dera (2007) and Dera et al. (2013), and data were ana- The sample was placed at the rotation center of the diffrac- lyzed using the GSE_ADA/RSV program. Integrated tometer and aligned using an optical microscope. A total peak intensities were corrected for Lorenz, polarization, angular range from φ = 56° to 125° (total angular opening of DAC absorption and sample displacement effects using ± 34.5°) was covered during the scans. A series of step and the methods implemented in GSE_ADA. Because of the wide-step φ-exposures were collected. Step scans involved high incident energy, low absorption coefficient and neg- 1° angular increments, while wide-step scans had 9.8° angu- ligible sample thickness the effects of sample absorption lar increments. The exposure time was at 3 s/°. After collec- were ignored. The structure of the β-phase was refined tion of step and wide-step φ-exposures at the zero detector using an initial cummingtonite model from Yang et al. position, more wide-step φ-exposures were recorded with (1998). The structure of the new high-pressure γ-phase at the detector rotated about its horizontal axis (2θ) by 20° 22.8(4) GPa was solved using the initial ambient pressure and then with the detector rotated about the vertical axis (ν) model from Finger (1969). Least-squares structure refine- by 10° and − 10°. Exposure time for the non-zero detector ment for selected pressures was done with the program position was at 6 s/°. SHELXL (Sheldrick 2008). The procedure for refinement The monochromatic diffraction experiment at 13ID-D was of the high-pressure data was similar to the ambient pres- conducted in a similar manner to those performed at 13-BM- sure data, however, all atoms in the high-pressure data C. X-rays with wavelength of 0.295 Å (42 keV) were used were refined with isotropic ADPs due to limited cover- with a focused X-ray beam size of 0.003 mm × 0.003 mm. age of reciprocal space. The site occupancies of all the Diffraction images were collected using a MAR165 charge high-pressure refinements were constrained to those deter- coupled device (CCD) detector, placed at a sample-to detec- mined from the ambient structure refinement. In some of tor distance of approximately 200 mm. The total rotation the high-pressure data, we were unable to locate hydrogen range around the vertical axis of the instrument (ω) was atoms, to keep the refinements consistent, we have opted ± 22°, with step scans covering 1° width and exposure time to have all hydrogen atoms omitted from the high-pres- at 0.5 s/°. sure structural refinement. Details of the crystal structure Step φ-exposures (13BM-C) and ω-scans (13ID-D) refinement, unit cell parameters at each pressure, refined were used in reconstruction of the crystal’s reciprocal lat- fractional coordinates for all the atoms, bond lengths and tice to determine the unit cell parameters and to index the atomic displacement parameters for selected pressures are diffraction pattern. Wide-step φ-exposures and ω-scans given in Tables 2, 3, 4 and 5. Unfortunately, only one pres- were used to determine d-spacings, azimuthal angles sure point of the new high-pressure phase produced data around the beam center and peak intensities of each dif- allowing to solve and refine the structure, due to limited fraction peak to solve the crystal structure. Data collection coverage of reciprocal space.

Table 2 Representative single- Run no. 0 1 3 3 crystal structure refinement for grunerite at selected pressures Beamline – 13BM-C 13BM-C 13BM-C Wavelength (Å) 0.560 0.434 0.434 0.434 Pressure (GPa) 0 1.13(2) 10.6(2) 22.8(4) Temperature (K) 298 298 298 298 θ range for data collection 3.077–25.547 1.501–19.542 1.401–13.796 1.446–23.182 No. of reflections collected 7782 920 669 2466 No. of independent reflections 1795 258 287 257 Reflections violating C2/m S.G 338 No. of parameters refined 103 42 84 42 Limiting indices − 14 ≤ h ≤ 14 − 14 ≤ h ≤ 8 − 4 ≤ h ≤ 4 − 6 ≤ h ≤ 6 − 28 ≤ k ≤ 28 − 8 ≤ k ≤ 7 − 19 ≤ k ≤ 18 − 24 ≤ k ≤ 24 − 8 ≤ l ≤ 5 − 6 ≤ l ≤ 6 − 5 ≤ l ≤ 5 − 7 ≤ l ≤ 7

Space group C2/m C2/m P21/m C2/m

Rint 0.0672 0.0611 0.0682 0.1218 Goodness of fit 1.083 1.109 1.139 1.267

wR2 0.0928 0.1204 0.1587 0.2406

R1 0.0382 0.0432 0.0592 0.0998

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 219

Table 3 Unit cell parameters of Run Pressure (GPa) a (Å) b (Å) c (Å) β (°) V (Å3) Space group grunerite at various pressures 0 0 9.553(1) 18.327(2) 5.3382(8) 101.854(4) 914.74(8) C2/m 1 1.13(2) 9.504(1) 18.24(1) 5.3224(8) 102.04(1) 902.6(8) C2/m 1 2.22(4) 9.445(1) 18.18(1) 5.3097(7) 102.38(1) 891.0(8) C2/m 1 3.63(7) 9.390(1) 18.11(1) 5.2765(8) 102.63(1) 875.8(9) C2/m 1 5.2(1) 9.338(1) 18.03(1) 5.2510(9) 102.86(1) 862.0(8) C2/m

1 7.4(1) 9.261(1) 17.92(1) 5.2196(8) 103.04(1) 844.2(9) P21/m

2 9.0(1) 9.226(4) 17.804(1) 5.1898(3) 103.15(1) 830.1(3) P21/m

3 10.6(2) 9.18(1) 17.751(2) 5.1694(7) 103.13(2) 820.8(5) P21/m

2 12.0(2) 9.107(6) 17.689(2) 5.1524(4) 103.34(2) 807.7(4) P21/m

2 16.3(3) 8.956(8) 17.497(2) 5.1040(5) 103.36(2) 778.2(5) P21/m 3 19.2(3) 9.332(1) 17.326(2) 4.9456(8) 107.52(3) 762.5(9) C2/m 2 20.6(4) 9.31(1) 17.298(5) 4.9210(9) 108.03(4) 753(1) C2/m 3 22.8(4) 9.287(6) 17.203(1) 4.8989(4) 107.99(1) 744.3(5) C2/m 2 23.0(4) 9.277(9) 17.1759(3) 4.8934(6) 108.25(3) 740.5(8) C2/m 2 25.6(5) 9.243(9) 17.127(4) 4.8706(6) 108.63(3) 730.7(8) C2/m

Results and discussion At 19.2(3) GPa, the structure adopts a C-centered lattice again, as determined through the analysis of systematic Phase transition in grunerite absences in the diffraction pattern. A similar high-pressure phase was previously indicated in a synthetic amphibole Three different phases were observed in grunerite on com- with composition, Na(NaMg)Mg5Si8O22(OH)2, by infrared pression to 25.6(5) GPa. A comparison of all three struc- spectroscopy experiments (Iezzi et al. 2006), based on the tures is shown in Fig. 1. The previously reported C2/m presence of a single OH-stretching band, however, until (α)–P21/m (β) transition was observed between 5.2(1) now there have been no previous diffraction experiments and 7.4(1) GPa and another transformation was detected reported to confirm this and constrain the crystal structure. between 16.3(3) and 19.2(3) GPa. Unit cell parameters of This new phase adopts a monoclinic structure with space grunerite up to 25.6(5) GPa are listed in Table 3. The new group C2/m, with unit cell parameters at 22.8(4) GPa, a b c phase transition transformed the symmetry from P21/m (β) = 9.287(6) Å, = 17.203(1) Å, = 4.89(1) Å, 3 to a previously unreported structure. This new structure β = 107.99(1)° and V = 744.4(5) Å . The β–γ phase tran- has monoclinic space group C2/m (γ), determined through sition in grunerite is accompanied by a discontinuous analysis of systematic absences in the diffraction pattern increase in the a unit cell parameter and β angle (Fig. 2). and, as evidenced through structure solution and refine- The a unit cell parameter of β-grunerite at 16.3(3) GPa is ment. The structure of all three phases have been solved 8.956(8) Å and in γ-grunerite at 19.2(3) GPa it increases and refined (Table 2). to 9.34(1) Å. Furthermore, β increases from 103.36(2)° At 7.4(1) GPa, the observed structure adopts a primi- to 107.52(1)° across the transition. Sueno et al. (1973) tive lattice based on reflections violating the C2/m space defined the tetrahedral displacement parameter ‘d’, as the group. The α–β phase transition is thus expected to occur distance between the centers of two opposing six-mem- between 5.2(1) and 7.4(1) GPa for the studied sample. The bered tetrahedral rings, and found a negative linear corre- examined crystal transformed to the P21/m β-phase by lation between d and β. Figure 3 shows a plot of calculated 7.4(1) GPa. There is a small slope change in a, b, c and β d values against β, and confirms this negative linear rela- at the C2/m–P21/m transition (Fig. 2). In all four unit cell tionship. Whittaker (1960) has associated the degree of the parameters the slope has decreased suggesting a change in closest packing of the tetrahedral chains with an increase the compression mechanism. Yang et al. (1998) estimated in β, which follows directly from Prewitt and Down’s 8th that the C2/m–P21/m transition pressure has a linear depend- rule of thumb on high pressure effects on bonding and ence on XFe with the relationship Ptr = − 1.23 + 4.52 XFe. coordination number (Prewitt and Downs 1998), that high- Based on this linear dependence, our sample would be pressure structures tend to be composed of closest-packed expected to transform to β-grunerite at ~ 2.21 GPa, which is arrays of atoms. much lower than our reported value. This suggests that the α–β phase transition pressure is affected by other factors in addition to XFe.

1 3 220 Physics and Chemistry of Minerals (2019) 46:215–227

Table 4 Atomic positional coordinates and atomic isotropic displace- Table 5 Bond lengths (Å) for grunerite at selected pressures ment parameters Pressure (GPa) 1.13(2) 10.6(2) 22.8(4) Pressure Atom X y z Ueq (GPa) Set A Set B

1.13(2) M1 0 0.0876(3) 0.5 0.010(1) M1-O1 2.075(6) 2.04(3) 2.01(3) 2.06(3) M2 0 0.1778(4) 0 0.010(1) M1-O2 2.12(1) 2.06(3) 2.02(2) 2.00(2) M3 0 0 0 0.009(1) M1-O3 2.112(7) 2.14(3) 2.02(2) 2.07(3) M4 0 0.2578(3) 0.5 0.010(1) Average 2.10233 2.04833 2.04333 Si1 0.2887(2) 0.0836(4) 0.2736(5) 0.006(1) M2-O1 2.14(1) 2.08(3) 2.05(2) 2.08(2) Si2 0.2993(2) 0.1680(4) 0.7783(5) 0.006(1) M2-O2 2.111(5) 2.04(3) 2.03(3) 1.96(2) O1 0.1151(5) 0.0882(8) 0.208(1) 0.008(1) M2-O4 2.06(1) 1.99(2) 2.03(2) 1.94(1) O2 0.1250(5) 0.1718(8) 0.715(1) 0.007(1) Average 2.10367 2.03667 1.99333 O3 0.1157(8) 0 0.705(1) 0.009(2) M3-O1 2.12(1) 2.02(3) 1.99(2) 2.05(2) O4 0.3824(6) 0.243(1) 0.767(1) 0.013(1) M3-O3 2.093(8) 2.12(4) 2.08(3) 2.08(3) O5 0.3512(5) 0.130(1) 0.064(1) 0.011(1) Average 2.1065 2.0525 2.065 O6 0.3513(5) 0.1174(8) 0.558(1) 0.010(1) M4-O2 2.15(1) 2.15(3) 2.05(3) 2.06(2) O7 0.3418(9) 0 0.268(1) 0.013(2) M4-O4 1.985(5) 2.02(2) 1.98(2) 1.94(2) 10.6(2) M1 − 0.2535(1) 0.3355(1) 0.4687(7) 0.009(1) M4-O5 2.45(1) 2.23(1) M2 − 0.252(1) 0.4257(2) 0.9697(7) 0.011(1) M4-O6 2.73(1) 2.26(2) 2.96(2) M3 − 0.251(1) 0.25 0.9702(9) 0.010(1) Average 2.28833 2.26714 2.07667 M4 − 0.2604(8) 0.5101(2) 0.4605(6) 0.012(1) T1-O1 1.616(6) 1.61(4) 1.66(4) 1.53(4) Si1A 0.044(3) 0.3341(3) 0.248(1) 0.007(2) T1-O5 1.61(1) 1.63(2) 1.65(2) 1.56(1) Si1B 0.547(2) 0.8318(3) 0.314(1) 0.009(2) T1-O6 1.631(9) 1.64(1) 1.63(1) 1.64(1) Si2A 0.040(3) 0.4207(3) 0.750(1) 0.011(2) T1-O7 1.608(7) 1.59(1) 1.61(2) 1.58(1) Si2B 0.558(2) 0.9152(3) 0.819(1) 0.009(2) Average 1.61625 1.6175 1.6375 1.5775 O1A − 0.136(6) 0.3357(6) 0.178(3) 0.013(5) T2-O2 1.622(6) 1.53(4) 1.68(4) 1.60(4) O1B 0.362(5) 0.8355(6) 0.234(3) 0.016(4) T2-O4 1.59(1) 1.61(2) 1.58(2) 1.58(1) O2A − 0.131(6) 0.4211(7) 0.685(3) 0.018(5) T2-O5 1.64(1) 1.67(2) 1.66(1) 1.66(1) O2B 0.370(5) 0.9201(6) 0.745(3) 0.013(4) T2-O6 1.64(1) 1.68(2) 1.63(2) 1.65(1) O3A − 0.115(6) 0.25 0.691(4) 0.014(5) Average 1.623 1.6225 1.6375 1.6225 O3B 0.373(5) 0.75 0.735(3) 0.004(4) O4A 0.122(3) 0.5015(6) 0.789(2) 0.005(3) Equation of state, bulk moduli and linear O4B 0.631(4) 0.9907(6) 0.750(2) 0.011(3) compressibilities O5A 0.117(4) 0.3662(6) 0.010(3) 0.015(4) O5B 0.616(4) 0.8948(6) 0.139(2) 0.009(3) O6A 0.117(4) 0.3855(6) 0.512(2) 0.014(3) Weighted volume and pressure data from all three phases O6B 0.608(5) 0.8513(6) 0.630(3) 0.013(3) were used to fit the second-order Birch–Murnaghan (BM) O7A 0.104(6) 0.25 0.305(3) 0.013(5) equation of state (EOS) using the program EOS-FIT V7 pro- O7B 0.611(6) 0.75 0.260(3) 0.009(5) gram (Gonzalez-Platas et al. 2016). The results of the EOS 22.8(4) M1 0 0.0857(2) 0.5 0.005(1) are plotted in Fig. 4 and shown in Table 6. The bulk moduli M2 0 0.1747(2) 0 0.004(1) determined from this study is in good agreement with Yang V M3 0 0 0 0.006(1) et al. (1998), the larger 0 in our study is likely due to greater VI 2+ M4 0 0.2638(2) 0.5 0.007(1) Fe content in our sample [ionic radius of Mg = 0.72 Å VI 2+ Si1 0.307(1) 0.0822(2) 0.340(1) 0.006(1) and high spin Fe = 0.78 Å (Shannon 1976)]. Si2 0.301(1) 0.1681(2) 0.833(1) 0.004(1) Linear compressibilities, defined as l0β = 1/3Kl0 (Angel O1 0.135(4) 0.0861(5) 0.236(3) 0.004(3) 2000), were determined by weighted least-squares fit of O2 0.121(4) 0.1724(5) 0.735(3) 0.008(3) the linearized second-order BM equation of state (Fig. 2; O3 0.139(5) 0 0.738(4) 0.002(3) Table 6). Linear compressibilities for α-grunerite are −1 O4 0.380(4) 0.2439(5) 0.759(3) 0.010(2) 0.0052(1), 0.0035(1) and 0.0038(1) GPa for βa, βb and O5 0.363(3) 0.1499(4) 0.184(3) 0.006(2) βc, respectively, with a ratio of 1.49:1.00:1.10. For the P m O6 0.368(3) 0.0946(5) 0.690(3) 0.009(3) 21/ polymorph the axial compressibilities were 0.007(1), −1 O7 0.363(5) 0 0.267(4) 0.009(3) 0.0038(3) and 0.00381(2) GPa for βa, βb and βc, respectively, with a ratio of 1.86:1.00:0.99. Zhang et al. (1992)

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 221

Fig. 1 (100) projection of the partial structure of grunerite show- β-grunerite phase, the reduction in symmetry causes the double- ing the structural changes across the α–β–γ phase transition. In the chains to split into two crystallographically distinct chains, the α and γ phase the double-chain of tetrahedra are O-rotated. In the A-chain (S-rotated) and the B-chain (O-rotated)

reported βa:βb:βc of 1.42:1.00:1.03 for a single-crystal X-ray All three phases display strong compressional anisotropy. study on natural grunerite up to 5.1 GPa, which is in good In the ambient pressure and P21/m phase the a axis is the agreement with our reported values for the ambient phase. most compressible while the b and c axis display similar For the new high-pressure γ-phase the linear compress- compressibilities. In the new high-pressure γ-phase the a ibilities were 0.0023(4), 0.0040(8) and 0.005(1) GPa−1 for axis is the least compressible while the most compressible βa, βb and βc, respectively, with a ratio of 0.56:1.00:1.41. direction is along the crystallographic c axis indicating a

1 3 222 Physics and Chemistry of Minerals (2019) 46:215–227

Fig. 2 Normalized unit cell parameters of grunerite from this study Fig. 4 Unit cell volume of grunerite from this study (black, blue and (black, blue and red) plotted against pressure. Results from previ- red), with a second-order BM EOS, compared to previous experi- ous experiments are shown in green (Zhang et al. 1992) and magenta ments [green and magenta, Zhang et al. (1992) and Yang et al. (Yang et al. 1998). Results of linearized second order BM EOS fit for (1998), respectively]. Circles, squares and triangles are the ambient each axis are shown with solid, dash-dot and dashed lines for a/a0 (in pressure α-phase, β-phase and high-pressure γ-phase, respectively black), b/b0 (in blue), c/c0 (in red), respectively

pressures. The difference between the kinking angle of the two chains (Δθ) decreases from 19.8(5)° to 16.0(5)° from 7.4(1) to 16.3(3) GPa. During compression the sense of rotation of both A and B chains remain the same, the A chain being S-rotated and the B chain O-rotated. The change in rotation type in the A chain to S-type parallels the clinopyroxene C2/c to P21/c transition where the A chain is also S-rotated and more extended than the B chain, which is O-rotated and significantly more kinked (Hugh-Jones et al. 1994). Yang et al. (1998) observed a change in the sense of rotation in the A chain from O to S-rotated with increasing pressure, however, our results show that the A-chain remains S-rotated throughout. In the new γ-phase of grunerite, the A and B chain consequently become one distinct chain Fig. 3 Relationship between β and the chain displacement factor in grunerite because of the change in symmetry, geometrically they are both equal to the B chain in the β-phase. Similarly, to the α-phase, the silicate chains in the high-pressure γ-phase are change in the compression mechanism. It should be noted also O-rotated. Of more importance, however, is the change that the degree of compressional anisotropy increases with in the kinking angle, which at 22.8(4) GPa is 137.5(4)°. each phase transition. In clinopyroxenes the HT C2/c structure is characterized by chains that are nearly fully extended, whereas the HP Structural changes with pressure C2/c phase displays tetrahedral chains that are more kinked (Arlt et al. 2000; Yang and Prewitt 2000; Tribaudino et al. The kinking angle of the silicate chains are characterized by 2001, 2003). Correspondingly to the structural changes the O5–O6–O5 angle. With increasing pressure, the kinking observed in pyroxenes, the low-pressure C2/m α-grunerite is angle in α-grunerite decreases from 171.3(1)° at ambient characterized by silicate chains which are slightly bent from pressure to 168(3)° at 5.2(1) GPa (Fig. 5). During the α–β being fully extended [171.3(1)° at ambient pressure], while phase transition, the silicate chain becomes two crystallo- the high-pressure C2/m γ-grunerite has silicate chains that graphically unique chains, the A and B chain. Upon further display more kinking [137.5(4)° at 22.8(4) GPa]. The primi- compression, the A and B chain kinking angle decreases. tive lattice, β-grunerite phase, is an intermediate structure The A chain kinking angle decreases from 166(4)° at having two silicate chains displaying different behavior from 7.4(1) GPa to 157(1)° at 16.3(3) GPa, while in the B chain both the low and high-pressure C-centered polymorphs. The it decreases from 146(4)° to 141(1)° at the same respective structural evolution of the double silicate chains is shown in

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 223

Table 6 Equation of state data for grunerite

References Phase Pressure medium Pmax(GPa) K0T (GPa) K′0T βa βb βc βa: βb: βc

Zhang et al. (1992) C2/m 4:1 meth-eth 5.1 50(1) 13(1) 0.00497(6) 0.00350(4) 0.0062(5) 1.41:1.00:1.03 Yang et al. (1998) C2/m 4:1 meth-eth 7.9 78(3) 4 0.0068(2) 0.0024(3) 0.0028(1) 2.83:1.00:1.17

Yang et al. (1998) P21/m 4:1 meth-eth 7.9 71(1) 6.1(5) 0.0043(3) 0.0029(1) 0.0030(1) 1.48:1.00:1.03 This study C2/m Ne 25.6(5) 78(1) 4 0.0052(1) 0.0035(1) 0.0038(1) 1.49:1.00:1.10

This study P21/m Ne 25.6(5) 66(4) 4 0.007(1) 0.0038(3) 0.00381(2) 1.86:1.00:0.99 This study C2/m Ne 25.6(5) 66(13) 4 0.0023(4) 0.0040(8) 0.005(1) 0.56:1.00:1.41

Composition: 2 + 3+ Zhang et al. (1992)—(Fe 5.33Mg1.46Fe 0.14Na0.05K0.01Al0.01)(Si7.92Al0.08)O22(OH1.92F0.05Cl0.01)

Yang et al. (1998)—(Fe3.272 Mg3.445Ca0.076Mn0.199Al0.008)(Si7.983Al0.017)O22(OH)2

This study—(Fe5.237Mg1.646Ca0.061Mn0.051Na0.015Ti0.002Cr0.001K0.001)(Si7.932Al0.083)O22(OH)2

Fig. 5 O5–O6–O5 kinking angle in grunerite as a function of pres- Fig. 6 Variation of M4–O5 distances in grunerite with pressure. sure. The kinking angle of the A chain (black squares) are plotted as Black squares are M4–O5A distances and blue squares are M4–O5B 360° minus the O5A–O6A–O5A angle to maintain the same analogy distances in the β-phase. Results from Yang et al. (1998) are shown as with clinopyroxenes. Circles, squares and triangles are the ambient magenta markers pressure α-phase, β-phase and high-pressure γ-phase, respectively. Blue squares are the B-chain in the P21/m phase. Results from Yang et al. (1998) are shown as magenta markers Yang et al. (1998) discussed the correlation between the variation of O5–O6–O5 angle and changes in M4–O5 and Fig. 1. It is interesting to note, as discussed by Papike and M4–O6 distances due to the C2/m-P21/m phase transition Ross (1970), that with further kinking towards the maxi- in cummingtonite. The values for M4–O5 and M4–O6 dis- mum angle of 120° the hexads of SiO4 tetrahedra will pos- tances are shown in Figs. 6 and 7. Our study is in good sess threefold rotation symmetry. Complete O-rotation of agreement with these observations. In the P21/m β-phase, the the double chains will result in a cubic close packing of M4–O5A distance increases with pressure from 3.50(4) Å at oxygen atoms. In the HP C2/c clinopyroxene structure, oxy- 7.4(1) GPa to 3.70(2) Å at 16.3(3) GPa, while the M4–O5B gen atoms also display behavior near that of cubic closest- distance decreases across the same pressure range from packing due to the extreme kinking of the silicate chains. 2.54(5) Å to 2.31(2) Å. This is due to the increase in coor- While the degree of closest packing in amphiboles can be dination from six to seven in the M4 site as the structure characterized by an increase in the monoclinic β angle this transitions from the α-phase to the β-phase. The kinking is not the case for clinopyroxenes. In previous high-pres- of the silicate chain pushes the O5B atom to coordinate sure experiments on clinopyroxenes, the monoclinic β angle with M4 while O5A moves further away. With increasing decreases with pressure and has a discontinuous decrease pressure as the structure changes from P21/m to the high- across the P21/c to HP C2/c phase transition (Hugh-Jones pressure C2/m γ-phase the M4–O6 distance is significantly et al. 1994; Arlt et al. 1998; Tribaudino et al. 2001; Alvaro increased to 3.00(1) Å due to further kinking of the tetra- et al. 2010), this is in contrast with our study where the β hedral chains. The coordination number in M4 decreases angle increases with pressure. to six across the phase transition, as the O6 atom moves

1 3 224 Physics and Chemistry of Minerals (2019) 46:215–227

to the M2 polyhedron in pyroxenes, which also displays similar properties. In the HP C2/c clinopyroxene phase, the extreme kinking of the silicate-chain involves breaking of bonds between O3 and M2 atoms, as a consequence, the M2 site no longer shares any edges with the silicate chain (Hugh-Jones et al. 1994; Downs 2003). In a similar manner, in γ-grunerite the extreme kinking of the double silicate chains leads to no sharing of edges between the SiO4 tetrahedra and the M4 polyhedron.

Implications

The close similarities of the physical, chemical and crys- Fig. 7 Variation of M4–O6 distances in grunerite with pressure. tallographic properties between amphiboles and pyroxenes Black squares are M4–O6A distances and blue squares are M4–O6B distances in the β-phase. Results from Yang et al. (1998) are shown as have been known for quite some time (Warren 1930; War- magenta markers ren and Modell 1930). Carpenter (1982) determined that the high-temperature to low-temperature displacive transformations in amphiboles and pyroxenes to be exactly analogous, further away from the coordination sphere, whereas the O5 even in the resulting microstructures. The non-ambient atom moves closer in Fig. 8. In the α-phase the M4 shares behavior of clinopyroxenes have been well studied across five edges with surrounding polyhedra (Fig. 9). Two edges a wide variety of compositions (Brown et al. 1972; Smyth are shared with the M2 polyhedra, one edge with the M1 1974; Hugh-Jones et al. 1994; Arlt and Armbruster 1997; polyhedron and two edges with T2 polyhedra. In the β-phase Arlt et al. 1998, 2000; Yang and Prewitt 2000; Tribaudino the M4 shares an additional edge with the T2A tetrahedron, et al. 2001, 2003; Nestola et al. 2008; Alvaro et al. 2010). decreasing the stability of the ionic structure due to the These studies have shown that clinopyroxenes undergo a increase in cation–cation repulsion as per Pauling’s third series of phase transformations from the high-temperature- rule. During the β–γ phase transition, as the M4 coordination C2/c to P21/c to high-pressure-C2/c phase. Based on their number decreases back to six, the stability of the polyhe- comparable behavior, a similar series of phase transitions is dral configuration increases, as the number of shared edges expected in clinoamphiboles. Our single-crystal experimen- decreases from six in β-grunerite, to three in γ-grunerite. In tal data have shown the existence of a new phase of grunerite the γ-phase, the M4 polyhedron shares two edges with the above 19.2(3) GPa. This study is the first structural report to M2 polyhedra and one edge with the M1 polyhedron. As show the existence of three polymorphs within an amphibole the M4 site in amphiboles are considerably more distorted group mineral, which closely mirrors the phase transition than the M1, M2 and M3 sites and are generally filled with sequence in clinopyroxenes as mentioned above. The exist- relatively larger cations, it is appropriate to compare them ence of the γ-phase of grunerite illustrates the corresponding

Fig. 8 Atomic coordination of the M4 cation in grunerite a α-grunerite at ambient pressure b β-grunerite at 7.4(1) GPa c γ-grunerite at 22.8(4) GPa. Dashed lines indicate non-bonding and distances between atoms that are greater than 3 Å

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 225

Fig. 9 M4 polyhedra configuration in grunerite.a α-grunerite at sharing edges with M1, 2 × M2, T1B, T2A and T2B polyhedra c ambient pressure, M4 polyhedron sharing edges with M1, 2 × M2 γ-grunerite at 22.8(4) GPa, M4 polyhedron sharing edges with M1 and 2 × T2 polyhedra b β-grunerite at 7.4(1) GPa, M4 polyhedron and 2 × M2 polyhedra structural relations and demonstrates that the parallel phase (Wallace and Green 1991; Welch and Graham 1992; Konzett transformation behavior is not only limited to temperature et al. 1997; Ernst and Liu 1998; Niida and Green 1999; as proposed by Carpenter (1982), but also includes pressure. Fumagalli and Poli 2005). The high-pressure, however, The high-temperature-C2/c to P21/c to high-pressure-C2/c may have an effect of increasing the dehydration tempera- transformations in clinopyroxenes (Arlt et al. 2000; Nestola ture. Constraining the stability of this metastable phase is et al. 2008) is analogous to the α to β to γ-phase transition important as it may have significant geophysical and petro- seen in this study. It is worth mentioning that in both clino- logical consequences, since amphiboles are commonly used pyroxenes and clinoamphiboles, high-pressure and high- as petrogenetic indicators and in geodynamic modelling. temperature phase transitions have the same space group Phases similar to γ-grunerite may exist for other clino- and and both phases are isometric to each other (Tribaudino et al. orthoamphiboles of different composition, therefore, fur- 2001, 2003). ther high-pressure investigations of these systems should be Equilibrium phase transformation sequences and chemi- encouraged. In addition, simultaneous high-temperature and cal reactions experienced by the major rock forming min- high-pressure studies on grunerite are needed to constrain erals have been extensively studied and are well under- the stability of γ-grunerite and to determine the dehydration stood. Metastable transformations, however, are poorly temperature of this phase. constrained. Constraining the stability of these metastable phases is important, as they may have significant geophysi- Acknowledgements The project was supported by the National Sci- ence Foundation Division of Earth Sciences Geophysics Grant No. cal implications as suggested by Agrusta et al. (2014) and 1722969. Portions of the X-ray diffraction work were conducted using Tetzlaff and Schmeling (2000) for both olivine and pyroxene the X-ray Atlas instrument at the University of Hawaii, funded by NSF in subducting slabs. In the case of both these minerals, meta- EAR Instrumentation and Facilities Grant 1541516. Portions of this stability promotes slab stagnation within the mantle transi- work were performed at GeoSoilEnviroCARS (Sector 13), Advanced Photon Source (APS), and Argonne National Laboratory. GeoSoilEn- tion zone due to the low-density metastable phases, which viroCARS is supported by the National Science Foundation—Earth provide positive buoyancy effects (Agrusta et al. 2014). It Sciences (EAR-1128799) and Department of Energy—Geosciences is conceivable that this metastable phase of grunerite would (DE-FG02-94ER14466). Use of the Advanced Photon Source was sup- exist in geologic environments such as the Tonga slab, where ported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. the thermal profile is lower than the mantle adiabat (Ganguly et al. 2009). It is estimated that the temperature of the Tonga slab within the mantle transition zone is less than 900 °C (Ganguly et al. 2009). The high-pressure and anomalously References cold-temperature of this region may be a likely geologic environment where metastable amphiboles like γ-grunerite Agrusta R, Hunen J, Goes S (2014) The effect of metastable pyrox- are preserved. The temperature in this region is near the ene on the slab dynamics. Geophys Res Lett 41:8800–8808 upper limit of amphibole stability before dehydration occurs

1 3 226 Physics and Chemistry of Minerals (2019) 46:215–227

Alvaro M, Nestola F, Ballaran TB, Cámara F, Domeneghetti MC, Iezzi G, Liu Z, Della Ventura G (2006) Synchrotron infrared spec- 2 Tazzoli V (2010) High-pressure phase transition of a natural troscopy of synthetic Na(NaMg)Mg5Si8O22(OH) up to 30 GPa: pigeonite. Am Miner 95:300–311 Insight on a new high-pressure amphibole polymorph. Am Miner Angel RJ (2000) High-pressure structural phase transitions. Rev Miner 91:479–482 A B Geochem 39:85–104 Iezzi G, Liu Z, Della Ventura G (2009) Synthetic Na (NaxLi1−xMg1 C Arlt T, Armbruster T (1997) The temperature-dependent P2 1/c-C2/c ) Mg5Si8O22(OH)2 (with x = 0.6, 0.2 and 0) P21/m amphiboles phase transition in the clinopyroxene kanoite MnMg[Si2O6]: a at high pressure: a synchrotron infrared study. Phys Chem Miner single-crystal X-ray and optical study. Eur J Miner 9:953–964 36:343–354 Arlt T, Angel RJ, Miletich R, Armbruster T, Peters T (1998) High- Iezzi G, Tribaudino M, Della Ventura G, Margiolaki I (2011) The high- pressure P2 1/c–C2/c phase transitions in clinopyroxenes: influence temperature P2 1/m → C2/m phase transitions in synthetic amphi- of cation size and electronic structure. Am Miner 83:1176–1181 boles along the richterite–(BMg)–richterite join. Am Miner 96:353 Arlt T, Kunz M, Stolz J, Armbruster T, Angel RJ (2000) P-T-X data Konzett J, Sweeney RJ, Thompson AB, Ulmer P (1997) Potassium on P2 1/c-clinopyroxenes and their displacive phase transitions. amphibole stability in the upper mantle: an experimental study in Contrib Miner Petrol 138:35–45 a peralkaline KNCMASH system to 8.5 GPa. J Petrol 38:537–568 Boehler R, De Hantsetters K (2004) New anvil designs in diamond- Law AD, Whittaker EJW (1980) Rotated and extended model struc- cells. High Press Res 24:391–396 tures in amphiboles and pyroxenes. Miner Mag 43:565–574 Boffa Ballaran T, Angel RJ, Carpenter MA (2000) High-pressure trans- Nestola F, Ballaran TB, Ohashi H (2008) The high-pressure C2/c–P21/c formation behaviour of the cummingtonite-grunerite solid solu- phase transition along the LiAlSi2O6–LiGaSi2O6 solid solution. tion. Eur J Miner 12:1195–1213 Phys Chem Miner 35:477–484 Brown G, Prewitt C, Papike J, Sueno S (1972) A comparison of the Nestola F, Pasqual F, Welch MD, Oberti R (2012) The effects of com- structures of low and high pigeonite. J Geophys Res 77:5778–5789 position upon the high-pressure behaviour of amphiboles: com- Carpenter M (1982) Amphibole microstructures: some analogies with pression of gedrite to 7 GPa and a comparison with anthophyllite phase transformations in pyroxenes. Miner Mag 46:395–397 and proto-amphibole. Miner Mag 76:987 Comodi P, Mellini M, Ungaretti L, Zanazzi PF (1991) Compressibility Niida K, Green DH (1999) Stability and chemical composition of par- and high pressure structure refinement of tremolite, pargasite and gasitic amphibole in MORB pyrolite under upper mantle condi- glaucophane. Eur J Miner 3:485–499 tions. Contrib Miner Petrol 135:18–40 Comodi P, Ballaran TB, Zanazzi PF, Capalbo C, Zanetti A, Nazzareni S Papike JJ, Cameron M (1976) Crystal chemistry of silicate minerals of (2010) The effect of oxo-component on the high-pressure behavior geophysical interest. Rev Geophys 14:37–80 of amphiboles. Am Miner 95:1042 Papike J, Ross M (1970) Gedrites-crystal structures and intracrystalline Dera P (2007) GSE-ADA data analysis program for monochromatic cation distributions. Am Miner 55:1945–1972 single crystal diffraction with area detector. GeoSoilEnviroCARS, Prewitt CT, Downs RT (1998) High-pressure crystal chemistry. Rev Argonne Miner 37:283–318 Dera P, Zhuravlev K, Prakapenka V, Rivers ML, Finkelstein GJ, Prewitt CT, Papike JJ, Ross M (1970) Cummingtonite: a reversible, Grubor-Urosevic O, Tschauner O, Clark SM, Downs RT (2013) nonquenchable transition from P2 1/m to C2/m symmetry. Earth High pressure single-crystal micro X-ray diffraction analysis with Planet Sci Lett 8:448–450 GSE_ADA/RSV software. High Press Res 33:466–484 Rivers M, Prakapenka VB, Kubo A, Pullins C, Holl CM, Jacobsen Dewaele A, Torrent M, Loubeyre P, Mezouar M (2008) Compression SD (2008) The COMPRES/GSECARS gas-loading system for curves of transition metals in the Mbar range: experiments and diamond anvil cells at the advanced photon source. High Press projector augmented-wave calculations. Phys Rev B 78:104102 Res 28:273–292 Downs RT (2003) Topology of the pyroxenes as a function of tempera- Shannon R (1976) Revised effective ionic radii and systematic stud- ture, pressure, and composition as determined from the procrystal ies of interatomic distances in halides and chalcogenides. Acta electron density. Am Miner 88:556–566 Crystallogr Sect A 32:751–767 Ernst WG, Liu J (1998) Experimental phase-equilibrium study of Al- Sheldrick G (2008) A short history of SHELX. Acta Crystallogr Sect and Ti-contents of calcic amphibole in MORB; a semiquantitative A 64:112–122 thermobarometer. Am Miner 83:952–969 Smyth JR (1974) The high temperature crystal chemistry of clinohy- Finger LW (1969) The crystal structure and cation distribution of a persthene. Am Miner 59:1069–1082 grunerite. Miner Soc Am Spec Pap 2:95–100 Sueno S, Cameron M, Papike J, Prewitt C (1973) The high temperature Fumagalli P, Poli S (2005) Experimentally determined phase relations crystal chemistry of tremolite. Am Miner 58:649–664 in hydrous peridotites to 6.5 GPa and their consequences on the Tetzlaff M, Schmeling H (2000) The influence of olivine metastabil- dynamics of subduction zones. J Petrol 46:555–578 ity on deep subduction of oceanic lithosphere. Phys Earth Planet Ganguly J, Freed AM, Saxena SK (2009) Density profiles of oceanic Inter 120:29–38 slabs and surrounding mantle: integrated thermodynamic and Thompson EC, Campbell AJ, Liu Z (2016) In-situ infrared spectro- thermal modeling, and implications for the fate of slabs at the scopic studies of hydroxyl in amphiboles at high pressure. Am 660 km discontinuity. Phys Earth Planet Inter 172:257–267 Miner 101:706 Gonzalez-Platas J, Alvaro M, Nestola F, Angel R (2016) EosFit7-GUI: Tribaudino M, Prencipe M, Nestola F, Hanfland M (2001) A P21/c– a new graphical user interface for equation of state calculations, C2/c high-pressure phase transition in Ca0.5Mg1.5Si2O6 clinopy- analyses and teaching. J Appl Crystallogr 49:1377–1382 roxene. Am Miner 86:807–813 Hawthorne FC, Oberti R (2007) Amphiboles: crystal chemistry. Rev Tribaudino M, Nestola F, Meneghini C, Bromiley G (2003) The high- Miner Geochem 67:1–54 temperature P2/c1–C2/c phase transition in Fe-free Ca-rich P2 1/c Hirschmann M, Evans BW, Yang H (1994) Composition and tempera- clinopyroxenes. Phys Chem Miner 30:527–535 ture dependence of Fe–Mg ordering in cummingtonite-grunerite Wallace M, Green DH (1991) The effect of bulk rock composition on as determined by X-ray diffraction. Am Miner 79:862–877 the stability of amphibole in the upper mantle: implications for Hugh-Jones D, Woodland A, Angel R (1994) The structure of high- solidus positions and mantle metasomatism. Miner Petrol 44:1–19 pressure C2/c ferrosilite and crystal chemistry of high-pressure Warren B (1930) II. The structure of tremolite H2Ca2Mg5(SiO3)8. Z C2/c pyroxenes. Am Miner 79:1032–1041 Kristalogr Cryst Mater 72:42–57

1 3 Physics and Chemistry of Minerals (2019) 46:215–227 227

Warren В, Modell D (1930) 11. The structure of anthophyllite Yang H, Prewitt CT (2000) Chain and layer silicates at high tempera- H2Mg7(SiO3)8. Z Kristalogr Cryst Mater 75:161–178 tures and pressures. Rev Miner Geochem 41:211–255 Welch MD, Graham CM (1992) An experimental study of glau- Yang H, Hazen RM, Prewitt CT, Finger LW, Ren L, Hemley RJ (1998) cophanic amphiboles in the system Na2O–MgO–Al2O3–SiO2–SiF4 High-pressure single-crystal X-ray diffraction and infrared spec- (NMASF): some implications for glaucophane stability in natural troscopic studies of the C2/m–P21/m phase transition in cumming- and synthetic systems at high temperatures and pressures. Contrib tonite. Am Miner 83:288–299 Miner Petrol 111:248–259 Zanazzi PF, Nestola F, Pasqual D (2010) Compressibility of proto- Welch MD, Cámara F, Ventura GD, Iezzi G (2007) Non-ambient in situ amphibole: a high-pressure single-crystal diffraction study of studies of amphiboles. Rev Miner Geochem 67:223–260 protomangano-ferro-anthophyllite. Am Miner 95:1758 Welch MD, Gatta D, Rotiroti N (2011) The high-pressure behavior of Zhang L, Ahsbahs H, Kutoglu A, Hafner SS (1992) Compressibility of orthorhombic amphiboles. Am Miner 96:623 grunerite. Am Miner 77:480–483 Whittaker E (1960) The crystal chemistry of the amphiboles. Acta Crystallogr A 13:291–298

1 3